5.2. ServerPeer attributes

When set to true, the server will stop if a database error occurs - e.g. database connection is lost - allowing other nodes to continue work. The default value is false.

The unique identifier of the ServerPeer. Each node deployed must have a unique identifier, whether the nodes form a cluster or are linked by a message bridge. The identifier must be a valid integer.

The default JNDI context to be used when binding queues. The default value is /queue.

The default JNDI context to be used when binding topics. The default value is /topic.

The post office used by the ServerPeer. You will not normally need to edit this attribute. The post office routes messages to queues and maintains the mapping between queues and addresses.

The default DLQ (Dead Letter Queue) that the server uses for destinations. You can override the DLQ on a per-destination basis. For more information about destinations, refer to Section 5.9, “Configuring Destinations”. A DLQ is a destination for messages that the server has failed to deliver more than a certain number of times. If the DLQ is not specified, the message will be removed after the maximum number of delivery attempts. You can specify a global default for the maximum number of delivery attempts with the DefaultMaxDeliveryAttempts attribute, or set the maximum individually on a per-destination basis.

The global default for the maximum number of times delivery will be attempted for a message before the message is removed or sent to the DLQ, if configured. The default value is 10. You can override this value on a per-destination basis.

The default expiry queue that the ServerPeer will use for destinations. You can override this value on a per-destination basis, as seen in the section on destination managed bean configuration. An expiry queue holds messages that have expired. Message expiry is determined by the value of Message::getJMSExpiration(). If the expiry queue is not specified, the message will be deleted when it expires.

This attribute lets you delay a redelivery attempt, which helps to prevent thrashing delivery-failure. The default value is 0 (that is, no delay). You can override this value on a per-destination basis.

This attribute defines the period of time between the server's queries to the queue for queue statistics. The default value is 5000 milliseconds.

The longest period (in milliseconds) that the client will wait for failover to begin on the server side when a problem is detected. The default value is 60000 (one minute).

The longest period (in milliseconds) that the client will wait for failover to complete on the server side once failover has been initiated. The default value is 300000 (five minutes).

JBoss Messaging provides a message counter history, which shows the number of messages arriving on each queue over a certain number of days. This attribute represents the maximum number of days for which to store message counter history. You can override this value on a per-destination.

The connection factory used to pull, or suck, messages between queues. You can omit this attribute to disable message sucking while retaining failover.

When true, JMS ordering is preserved in the cluster. See Chapter 6, Clustering Notes for more detail. The default value is false.

When failover occurs, messages that have been delivered will be stored while the clients reconnect. If the clients do not reconnect (for example, if the client is dead), these messages will eventually time out and be added to the queue. This attribute sets the period before timeout in milliseconds. The default value is 300000 (five minutes).

When set to true, enables message counters upon server start.

JBoss Messaging internally creates connections between nodes to redistribute messages between clustered destinations. These connections are created with a special, reserved username. This attribute defines the password to use when creating these connections.

For versions of JBoss Messaging later than 1.4.1.GA, you must define the SuckerPassword on the SecurityMetadataStore.

This is the maximum number of times a sucker's connection is permitted to retry in the event of a failure. The default value is -1 which represents "retry indefinitely".

This is the interval in milliseconds between each retry of the failed sucker's connection. The default value is 5000.

To enable strict JMS Technology Compatibility Kit (TCK) semantics, set this attribute to true.

Returns a list of the destinations (queues and topics) currently deployed.

A message counter for a particular queue.

Statistics about each message counter for each queue.

When this attribute is false, server-side failover does not occur when a node crashes in a cluster.

The persistence manager used by the ServerPeer. (You will not normally need to change this attribute.)

The JMS user manager used by the ServerPeer. (You will not normally need to change this attribute.)

The pluggable SecurityStore. If you redefine this attribute, remember that you will need to authenticate the MessageSucker user (JBM.SUCKER) with all special permissions required by clustering.

Specifies whether the transaction creation time is stored in the transaction record. If set to true, the transaction record is stored. The default is false.